Differences in spinal axial rotation angles during the lumbar-locked rotation test between hyper and normal thoracic rotation groups: A cross-sectional study

Abstract

BACKGROUND:

Trunk rotation is important in many sporting activities The thoracic spine has reciprocal relationships with the lumbar and pelvic spines, such that reduced flexibility in the lumbar or thoracic spine can lead to abnormal patterns of trunk movement and pain. However, few studies have investigated the relative trunk rotation mobilities of the thorax, lumbar, and pelvis.

OBJECTIVE:

To compare thoracic, lumbar, and pelvic rotation angles during the lumbar-locked rotation test between hyper and normal thoracic rotation groups.

METHODS:

Thirty-two young, active participants were enrolled in this study. After the attachment of inertial measurement units at the T1, T7, T12, L3, and S2 levels, the participants were required to stand in a comfortable upright posture for 5 s to allow postural measurements before performing the lumbar-locked rotation test. The participants were then divided into hyper thoracic rotation and normal thoracic rotation groups based on T1 angle measurements obtained during the lumbar-locked rotation test.

RESULTS:

The hyper thoracic rotation group had significantly higher thoracic rotation angles on both the right ( 0.05) and left ( 0.05) sides compared with the normal thoracic rotation group. Furthermore, we observed flat lumbar lordosis in the hyper thoracic rotation group compared with the normal thoracic rotation group, particularly in the lower lumbar region in standing posture.

CONCLUSION:

Our data suggest that evaluations of thoracic mobility should consider relative thoracic, lumbar, and pelvic motions, rather than the T1 angle alone. This study provides a basis for health professionals to evaluate movement dysfunctions associated with thoracic hypermobility.

1. Introduction

Trunk rotation is important in many sporting activities (e.g., swimming, tennis, baseball, and golf) and functional daily activities such as reaching, walking, and running [1, 2, 3]. Approximately 80% of axial rotation originates from the thoracic spine, with decreased range of motion in lower segments of the thoracic and lumbar spine because of the directions of facet joints [4, 5]. However, the thoracic spine has reciprocal relationships with the lumbar, pelvic, and cervical spines, such that reduced flexibility in the lumbar or thoracic spine can lead to abnormal patterns of trunk movement, pain, or injury [6, 7, 8]. Thus, continuous evaluation of relative trunk flexibility is necessary to prevent injuries and develop rehabilitation programs for athletes who must perform repetitive and extreme trunk rotational movements.

Many studies have evaluated trunk rotation flexibility in healthy people or athletes adopting various postures, including a half-kneeling, seated position, as well as forearm and kneeling postures, when using the lumbar-locked rotation test [9, 10, 11]. Additionally, the intra- and interrater reliabilities of thoracic spine rotation measurements have been quantified in competitive athletes and the general population [3, 12, 13]. Johnson et al. [9] reported good intra- and interrater reliabilities for thoracic spine rotation measurements obtained in sitting and lumbar-locked positions. Furthermore, some studies compared thoracic rotation angles between athletes and healthy controls. Furness et al. [12] assessed rotation measurements during the lumbar-locked rotation test in a group of elite male surfers; they found that the mean thoracic rotation angle was significantly higher (by $\sim$20^∘) among the surfers compared with a control group, although there was no group difference in the thoracic range of motion in the sagittal plane. These results may be explained by the repeated trunk rotation associated with many sports activities that require extreme trunk rotation; individuals engaging in such activities generally must acquire increased trunk mobility compared with the general population.

On the one hand, Harris-Hayes et al. [14] posited that suboptimal function in one part of the spine affects the mechanics of adjacent regions. Moreover, small loads in adjacent regions may contribute to tissue stress that, results in tissue injury over time. Many studies have established that reduced thoracic mobility can impair the functioning of anatomically related regions (such as the lumbar and cervical spine and shoulder) and is a predisposing factor for pain and injury [15, 16, 17]. Given the results of previous studies, one could presume that relative stiffness of the lumbar spine can lead to compensatory movements of joints in adjacent regions (i.e., thoracic or pelvic regions). In other words, if there is a relative increase in thoracic rotation movement, it implies that there may be a corresponding decrease in posture or movement in adjacent regions. Additionally, it suggests that there is a higher likelihood of experiencing pain in the areas where movement has relatively increased. Therefore, static posture assessment and evaluating the movement of the upper and lower thoracic spine, lumbar spine, and pelvis between the group with increased rotation and the normal group during the lumbar-locked test is an essential aspect of the assessment for thoracic function. However, few studies have analyzed the effects of increased thoracic rotation mobility on such movements. Furthermore, although many studies have investigated the reliabilities of thoracic rotation measurements, those studies mostly focused on the T1 level [11, 12, 13]; few studies have investigated relative trunk rotation mobilities of the thorax, lumbar, and pelvis.

Among the various non-invasive methods for quantifying trunk axial rotation, the goniometer [9, 11] or inclinometer [3, 12] are preferred by many movement experts. Both instruments are easy to use, cost-effective, reliable, and appropriate for non-clinical settings [9, 11, 12]. However, these instruments have limited utility for evaluations of the entire trunk and tend to be used to assess a single spinal region. Thus, three-dimensional motion analysis is required to accurately measure trunk rotation [18]. Three-dimensional motion analysis systems using reflective markers allow for highly accurate static and dynamic movement analysis, but they are expensive and require large amounts of space [18]. Inertial measurement units (IMUs) are considered reliable for evaluations of posture, trunk motions, movement dysfunction, and gait. These small, portable sensors are accurate, easy to use, and can monitor movements in various spinal segments [19, 20, 21]. Therefore, the purpose of this study was to compare thoracic, lumbar, and pelvic rotation angles in hyper and normal thoracic rotation groups during the lumbar-locked rotation test. We hypothesized that during the lumbar locked test, the hyper rotation angle group would tend to have greater thoracic rotation angles and smaller lumbar lordosis in the standing posture.

2. Materials and methods

2.1. Participants

We recruited 32 young asymptomatic adults (11 men and 21 women) for this cross-sectional study through advertisements placed around university campuses, in the Republic of Korea. None of the participants had experienced surgery, traumatic injury, or musculoskeletal disease during the 6 months prior to study enrollment; they were also free from functional restrictions, respiratory and neurologic disorders, and pain in the spine and lower limbs. Table 1 summarizes the participant characteristics. Ethical approval for this study was obtained from the University Ethics Committee for Human Investigations (Approval number 2020-07-016-001), and written informed consent was obtained from all participants.

Table 1.

General characteristics and standing angles of the participants. ($N=$ 32)

Variables	Hyper thoracic rotation group ($n=$ 16)	Normal thoracic rotation group ($n=$ 16)	$p$
Age (years)	24.1 $\pm$ 2.6	24.8 $\pm$ 2.5	0.410
Height (cm)	165.1 $\pm$ 7.0	167.6 $\pm$ 7.1	0.336
BMI (kg / m2)	22.1 $\pm$ 2.7	23.0 $\pm$ 2.7	0.373
Rt. Hamstring (∘)	26.6 $\pm$ 11.0	30.6 $\pm$ 12.5	0.345
Lt. Hamstring (∘)	25.6 $\pm$ 12.6	29.6 $\pm$ 14.2	0.408
Upper trunk extensor (kg)	65.3 $\pm$ 7.3	64.9 $\pm$ 10.3	0.911
Thoracic angle (∘)
Global	$-$32.9 $\pm$ 9.2	$-$38.8 $\pm$ 6.5	0.066
Upper	$-$28.5 $\pm$ 6.7	$-$32.1 $\pm$ 7.2	0.153
Lower	$-$4.5 $\pm$ 7.1	$-$6.2 $\pm$ 6.0	0.453
Lumbar angle (∘)
Global	14.8 $\pm$ 9.1	22.4 $\pm$ 10.5	0.041*
Upper	8.6 $\pm$ 8.1	7.9 $\pm$ 6.2	0.762
Lower	0.6 $\pm$ 5.4	15.3 $\pm$ 8.8	0.013*

All values are mean $\pm$ standard deviation. Abbreviation: BMI, body mass index. Hyper thoracic rotation group (T1 rotation angle during Lumbar-locked rotation test) 50^∘$⩽$ T1 70^∘; Normal thoracic rotation group, 30^∘$⩽$ T1 50^∘. Thoracic kyphosis global, T1 relative T12; Upper thoracic, T1 relative T7; Lower thoracic, T7 relative T12, Lumbar lordosis global, T12 relative S2 angle; Upper lumbar lordosis, T12 relative L3; Lower lumbar lordosis, L3 relative S2. ^* p< 0.05.

2.2. Inertial measurement units

We used five wireless inertial measurement units (IMUs) to measure trunk axial rotation during the lumbar-locked rotation test. Each IMU device comprises transmitters (EBIMU24G; E2BOX, Seoul, South Korea), a receiver (EBRF24G3CH; E2BOX), three gyroscopes, three accelerometers, three magnetometers, and Kalman filters. The transmitters measured 32 $\times$ 21 $\times$ 6.5 mm and weighed 7.85 g, including the attached lithium polymer battery. The IMU sensor radio-frequency data were processed using Visual FoxPro software (Microsoft Corp., Redmond, WA, USA), with a sampling frequency of 100 Hz. The sensitivities of the sensors were as follows: gyroscope, 250–2,000 dps; accelerometer, 2–8 g; and magnetometer, 1.3–8.1 gauss. The static accuracy of the IMUs was 0.5^∘, and the dynamic accuracy was 2^∘ [21].

The Eulerian coordinate system (i.e., roll–pitch–yaw) [21] was used to measure sagittal angles at the T1, T7, T12, L3, and S2 vertebral levels during standing. Each angle is reported as the mean of three 5-s measurements obtained during standing. The global thoracic kyphosis, upper thoracic, and lower thoracic angles were measured between T1 and T12, T1 and T7, and T7 and T12, respectively; the global lumbar lordosis (GLL), upper lumbar (UL${}_X)$, and lower lumbar angles (LL${}_X)$ were measured between T12 and S2, T12 and L3, and L3 and S2, respectively.

GLL is the angle between two intersecting lines, one indicating the inclination of the sensor at T12 and the other indicating the inclination of the sensor at S2. A positive lumbar angle indicates lumbar lordosis. The UL_X is the angle between two intersecting lines, one indicating the inclination of the sensor at T12 and the other indicating the inclination of the sensor at L3. The LL_X is the angle between two intersecting lines, one indicating the inclination of the sensor at L3 and the other indicating the inclination of the sensor at S2. Additionally, based on previous research and IMU system measurements, participants with a GLL angle less than 20^∘ were assumed to have reduced lumbar lordosis [21].

The Eulerian coordinate system was also used to measure rotation angles at the T1, T7, T12, L3, and S2 vertebral levels during the lumbar-locked rotation test. The participants were categorized into hyper rotation (T1 rotation angle $=$ 50–69^∘) and normal rotation (T1 rotation angle $=$ 30–49^∘) groups based on the T1 rotation mean value (49.8^∘) obtained during the lumbar-locked rotation test. Mean angles were calculated using Microsoft Excel 2010 software (Microsoft Corp.).

Participants performed the lumbar-locked thoracic rotation test for 13 s on each side (start position 3 s; maximal thoracic rotation, 5 s; hold, 5 s). The Eulerian angles were measured in the starting and maximal trunk axial rotation positions. Reliability indices for the wireless IMU system included intraclass correlation coefficients (ICCs) and 95% confidence intervals (CIs) (Table 2).

Table 2.

The intra-class correlation coefficient for IMU data during standing and trunk rotations during lumbar locked rotation test. ($N=$ 32)

Variables	Standing	Trunk rotation right side	Trunk rotation left side
	Sagittal plane	Coronal plane	Coronal plane
T1	0.984	0.962	0.945
T7	0.996	0.960	0.955
T12	0.995	0.981	0.975
L3	0.998	0.964	0.975
S2	0.999	0.925	0.980

2.3. Trunk extension strength

A digital handheld dynamometer (Power Track II; JTech Medical, Salt Lake City, UT, USA) was used to measure the maximum voluntary upper thoracic strength [22]. The erector spinae muscles act as rotators of the thoracic spine [23]. We measured trunk extensor muscle strength to ensure that differences between the two groups in trunk rotation angle were not attributable to variation in this parameter. The participants were instructed to lie prone, with the hips and knees extended and the arms at the side of the body. A padded transducer head was placed between the scapulae across the midline. Isometric strength was measured in one warm-up trial, followed by three successive 5-s maximum-effort trials [24]. The mean peak value of the three trials was calculated.

2.4. Hamstring length

An active knee extension test was conducted to evaluate the lengths of both hamstring muscles using a digital dual inclinometer (Acumar; Lafayette Instrument Co., Lafayette, IN, USA). A pressure biofeedback unit was used to prevent compensatory movements in the lumbopelvic region. Each participant was placed in the supine position, and the pressure biofeedback unit was placed between the examination bed and the lumbopelvic region. The measured leg was then flexed at 90^∘ at the hip and knee on an adjustable support table; the contralateral leg was fixed to the bed with a strap placed across the thigh. In this position, the plastic bag of the biofeedback unit was inflated to a pressure of 40 mmHg [25]. An inclinometer was positioned at the anterior tibial border, halfway between the inferior pole of the patella and a line connecting the two malleoli. A second inclinometer was placed on the anterior side of the thigh, 10 cm proximal to the superior pole of the patella. The lower leg was then extended while the participant relaxed the ankle. Active knee extension was performed until the participant felt resistance in the stretched hamstring muscle; at this point, measurements were recorded using the inclinometer. The active knee extension angle was defined as the angle of knee extension [26].

2.5. Procedures

We adapted the protocol for the lumbar-locked rotation test to measure participants’ thoracic rotation mobility, as in previous studies [3, 10]. The participants were asked to stand in a comfortable position while looking straight ahead, with their feet shoulder-width apart. The examiner marked the T1, T7, T12, L3, and S2 spinous processes of each participant, and the five transmitters were mounted on a plastic frame. The mounted transmitters were then attached to the T1, T7, T12, L3, and S2 spinous processes using medical tape (Transpore; 3M Korea Ltd., Seoul, South Korea), and data were acquired [10]. Then, the participants were asked to assume the four-point kneeling position; they were asked to maintain upper extremity support by placing the elbows and forearms in contact with their knees and aligning the most prominent point on the lateral side of their knee with the lateral line of their elbow while keeping the forearms straight. Subsequently, each participant was instructed to grasp their neck and slowly rotate the thoracic spine, without extending the lumbar spine or allowing the buttocks to come off the feet. While grasping their neck, each participant was instructed to keep the head aligned with the rotation of the thoracic spine. The participants remained in the starting position for 3 s (Fig. 1A), performed maximal trunk axial rotation for 5 s, and then held that position for an additional 5 s (Fig. 1B); a metronome was used during this sequence, which was repeated three times on each side after two practice trials. The Eulerian angles were measured at the T1, T7, T12, L3, L5, and S2 levels in the starting and maximal trunk axial rotation positions. Trials were regarded as failures if any of the following characteristics were observed: inability to assume the four-point kneeling position because of insufficient pelvic, hip, or knee flexion; loss of lumbar spine alignment; scapular retraction; or loss of upper extremity alignment, either unilaterally (because of an insufficient elbow angle or inability to keep the hand on the back of the neck) or bilaterally (because of an inability to keep the contralateral arm on the table) [9, 10].

Figure 1.

Lumbar-locked rotation test A) start position, B) end position.

2.6. Statistical analyses

Data were analyzed using SPSS for Windows software (version 22.0; IBM Corp., Armonk, NY, USA). The Kolmogorov–Smirnov test was used to assess the normality of the data, and $p$-values $>$ 0.05 were considered indicative of statistical significance. Demographic data, thoracic and lumbar spine angles, and pelvic angles during the lumbar-locked rotation test were compared between groups using independent $t$-tests (normally distributed variables). Despite the small sample size, the statistical test results indicated that the data followed a normal distribution, and the primary objective of the study was not to examine the median value for the rotation angle of the groups. Therefore, an independent $t$-test was utilized for verification. Effect sizes for observed differences were calculated by dividing the difference between the mean values of both groups by the combined standard deviation, and effect size strengths were interpreted using the guidelines of Cohen [27].

3. Results

The Kolmogorov–Smirnov test showed that the data were normally distributed ($p>$ 0.05).

3.1. Spinal sagittal angles in standing

Flat lumbar lordosis in the GLL (14.8^∘$\pm$ 9.1^∘; $p=$ 0.041) was observed in the hyper rotation angle group, particularly in the LL_X (8.6^∘$\pm$ 5.4^∘; $p=$ 0.013) in standing posture. These findings contrasted with the normal thoracic rotation group (22.4^∘$\pm$ 10.5^∘ and 15.3^∘$\pm$ 8.8^∘ for GLL and LL_X, respectively) (Table 1).

3.2. Lumbar-locked rotation test

Descriptive statistics for T1, T7, T12, L3, and S2 angles during the lumbar-locked rotation test are summarized in Table 3. During the lumbar-locked rotation test, the T1 rotation angles were significantly higher in the hyper thoracic rotation group than in the normal thoracic rotation group (58.9^∘$\pm$ 5.9^∘ and $-$51.7^∘$\pm$ 7.8^∘ vs. 40.7^∘$\pm$ 3.3^∘ and -37.2^∘$\pm$ 9.5^∘ on the right and left sides, respectively; both 0.001); such differences were also observed for T7 rotation angles (32.1^∘$\pm$ 4.2^∘ and $-$34.6^∘$\pm$ 8.2^∘ vs. 21.4^∘$\pm$ 5.6^∘ and $-$23.6^∘$\pm$ 9.4^∘ on the right and left sides; 0.001 and $p=$ 0.001, respectively) and T12 rotation angles (19.8^∘$\pm$ 5.0^∘ and $-$22.7^∘$\pm$ 4.7^∘ vs. 10.7^∘$\pm$ 5.4^∘ and $-$13.7^∘$\pm$ 8.4^∘ for the right and left sides; 0.001 and $p=$ 0.001, respectively).

Table 3.

Comparison of maximal trunk axial rotation angle during lumbar-locked rotation test between groups. ($N=$ 32)

Variables	Hyper thoracic rotation group ($n=$ 16)	Normal thoracic rotation group ($n=$ 16)	Effects size	$p$
T1 (∘)
Right side	58.9 $\pm$ 5.9	40.7 $\pm$ 3.3	3.970	0.001*
Left side	$-$51.7 $\pm$ 7.8	$-$37.2 $\pm$ 9.5	1.718	0.001*
T7 (∘)
Right side	32.1 $\pm$ 4.2	21.4 $\pm$ 5.6	2.230	0.001*
Left side	$-$34.6 $\pm$ 8.2	$-$23.6 $\pm$ 9.4	1.293	0.001
T12 (∘)
Right side	19.8 $\pm$ 5.0	10.7 $\pm$ 5.4	1.808	0.001*
Left side	$-$22.7 $\pm$ 4.7	$-$13.7 $\pm$ 8.4	1.367	0.001
L3 (∘)
Right side	10.0 $\pm$ 4.1	7.0 $\pm$ 4.3	0.727	0.056
Left side	$-$12.6 $\pm$ 3.4	$-$9.2 $\pm$ 6.2	0.713	0.060
S2 (∘)
Right side	6.6 $\pm$ 4.0	4.8 $\pm$ 3.9	0.468	0.209
Left side	$-$9.8 $\pm$ 5.2	$-$6.9 $\pm$ 6.7	0.504	0.178

All values are mean $\pm$ standard deviation. Hyper thoracic rotation group (T1 rotation angle during lumbar-locked rotation test) 50^∘$⩽$ T1 70^∘; Normal thoracic rotation group, 30^∘$⩽$ T1 50^∘. Effect size, Cohen’s d.

4. Discussion

In this study, we compared thoracic, lumbar, and pelvic angles in normal and hyper thoracic rotation angle groups during the lumbar-locked rotation test. All thoracic rotation angles (i.e., TI, T7, and T12) were higher in the hyper thoracic rotation group than in the normal thoracic rotation group. Furthermore, we observed flat lumbar lordosis in the hyper thoracic rotation group compared with the normal thoracic rotation group, particularly in the LL_X in standing posture, despite finding no differences between groups in trunk extensor strength or hamstring length.

Optimal spinal posture during daily activities is widely considered important in minimizing stress, deformation, and energy expenditure while maximizing mechanical advantage. Furthermore, the evaluation of posture and spinal mobility is important in clinical practice because posture plays a key role in influencing movement patterns. A previous review reported that the mean thoracic kyphotic angle of individuals in their twenties is 34^∘ (ranging from 29^∘ to 45^∘) when measured from C7 or T1 to T12 in a standing position using non-radiological devices [28]. In this study, we measured the thoracic sagittal angles of participants in standing posture, from T1 to T12, using an IMU. The thoracic kyphotic angles were $-$32.9^∘$\pm$ 9.2^∘ in the hyper rotation angle group and $-$38.8^∘$\pm$ 6.5^∘ in the normal thoracic rotation group. Both groups fell within the normal range, and there was no statistically significant difference between them. However, we observed flat lumbar lordosis in the GLL (14.8^∘$\pm$ 9.1^∘) in the hyper rotation angle group, particularly in the LL_X (8.6^∘$\pm$ 5.4^∘) in standing posture. González-Sánchez et al. [29] measured lumbar lordosis (L1–L5) in normal-weight participants using an electromagnetic tracking device and reported a mean lumbar lordosis angle of 28^∘. Furthermore, Shin and Yoo [21] used an IMU system to investigate lumbar postures in three groups based on GLL and regional lumbar lordosis (i.e., UL_X and LL${}_X)$ angles in standing posture. The authors reported that individuals with a GLL measurement of less than 20^∘ had a flat lumbar posture based on IMU measurements. Additionally, the flat lumbar group demonstrated angles of less than 10^∘ in both the UL_X and LL_X.

The lumbar-locked rotation test is useful for assessing coordinated trunk rotation and flexibility, particularly when monitoring effective rehabilitation [3, 9]. Moreover, this test specifically evaluates thoracic rotation mobility by minimizing the contributions of hip, pelvis, and lumbar spine motion during thoracic rotation. In this study, all thoracic rotation angles (i.e., TI, T7, and T12) were higher in the hyper thoracic rotation group than in the normal thoracic rotation angle group during the lumbar-locked thoracic rotation test using IMU measurement. The T1 angles of the normal rotation group were 40.7^∘ and $-$37.2^∘ on the right and left sides, respectively. Similarly, Johnson et al. [9] reported T1 angles of 39.0^∘ and 46.4^∘ on the right and left sides, respectively, among 46 healthy volunteers during the lumbar-locked rotation test. Whereas we used an IMU for measurements, Johnson et al. [9] used a bubble inclinometer during the lumbar-locked rotation test.

In the present study, the hyper rotation group exhibited T1 rotation angles of $-$58.9^∘ on the right side and $-$51.7^∘ on the left side. These results indicate that T1 rotation angles in the hyper rotation group are larger than such angles in normal individuals but smaller than the angles observed in sports players. Previous studies of the lumbar-locked rotation test in athletes have revealed T1 rotation angles ranging from 60^∘ to less than 70^∘ [3, 12]. In a study of competitive swimmers by Feijen et al. [3], good to excellent intra- and interrater reliabilities were observed for the lumbar-locked rotation test. The T1 angles on the right and left sides were 68.3^∘ and 63.5^∘, respectively, as assessed by the first evaluator. The second evaluator reported angles of 66.9^∘ and 63.0^∘ for the right and left sides, respectively. Moreover, Furness et al. [12] reported mean T1 angles of 63.1^∘ on the right side and 64.0^∘ on the left side during the lumbar-locked rotation test in elite male surfers compared with control group measurements of 41.5^∘ on the right side and 40.3^∘ on the left side. In the present study, the hyper thoracic rotation group of young healthy people exhibited smaller T1 angles than elite athletes, perhaps because elite athletes are required to perform repetitive and extreme trunk rotations during their physical activities.

We observed flat lumbar lordosis in the GLL in the hyper thoracic rotation group compared with the normal thoracic rotation group, particularly in the LL_X in standing posture, despite finding no differences between groups in trunk extensor strength or hamstring length. We believe that these results can be attributed to movement caused by the participants’ flat backs. Ideally, kinematic analysis of the spine should consider the whole trunk, rather than the first thoracic spine alone. During sports and daily activities, there is mechanical interaction among the thoracic, lumbar, and pelvic regions. Limited mobility in one region can lead to the accumulation of tissue stress and motion difficulties in adjacent anatomical regions [12, 30]. Previous findings regarding individuals with reduced lumbar lordosis, partially support our findings. Shin and Yoo [31] demonstrated upper trunk acceleration (T7) during gait in young individuals with reduced lumbar lordosis. The authors reported significantly higher upper trunk acceleration in anterior–posterior and vertical directions in the reduced lumbar lordosis group than in the normal lumbar lordosis group. Directional acceleration refers to the rate at which directional motion velocity changes over time. An increase in acceleration indicates that directional velocity undergoes frequent changes during walking [32]. Although the previous study [30] did not directly present the mobility of thoracic rotation, it can be considered that there was increased mobility of T7 during gait in the group with decreased lumbar lordosis compared to the normal lumbar group.

Additionally, repetitive and excessive rotation movement is a predisposing factor for postural instability, pain, and injury. In clinical practice, it is important to consider that the presence of a high range of motion in a mobility test may indicate a higher likelihood of pain in that region, as well as the possibility that restricted movement in adjacent body regions is a contributing factor. In the present study, the hyper rotation mobility of T1 in the hyper rotation group was possibly associated with an increased risk of low back pain and reduced mobility in the shoulder and neck regions. These results suggest a need for further evaluation and highlight the importance of additional research in this area. Heo et al. [33] found that lumbar stabilization exercises combined with thoracic mobilization stabilized the lumbar region, alleviated pain, and improved function through enhanced control of segment motion in chronic low back pain patients. Thus far, no studies have examined the role of relative thoracic hyper rotation movement during the lumbar-locked rotation test in managing impaired movement. Therefore, reverse conditions (i.e., the effect of flat back on thoracic movement) require further investigation.

In the present study, no differences in lumbar and pelvic angles were observed between the normal and hyper thoracic rotation angle groups during the lumbar-locked rotation test. The four-point kneeling position adopted in the lumbar-locked rotation test places the hips and lumbar spine into maximal flexion, which reduces the contributions of the pelvis and lumbar spine to thoracic rotation [9]. The criteria used in our study for a failed test included an inability to assume the quadruped position because of a loss of pelvis, hip, or knee flexion, along with the loss of lumbar spine alignment. Although the majority of previous studies using the lumbar-locked rotation test did not measure the lumbar or pelvic angles, they controlled for compensatory lumbar and pelvic movements to isolate T1 motion. Therefore, we believe that the lack of differences between groups in lumbar and pelvic angles can be attributable to the control of compensatory movement during the lumbar-locked rotation test.

This study had some limitations. First, the participants were all healthy individuals in their twenties without any pain and without spinal mobility limitations. The findings cannot be generalized to athletes or patient populations. The participants were all young, healthy, and asymptomatic. Future studies should examine trunk axial rotation in patients with thoracic or low back pain during the lumbar-locked rotation test. Second, we did not control for participation in sporting and other recreational activities, which could influence movement. Third, when measuring lumbar region kinematics with surface-based measurement systems, some measurement error is to be expected, because of skin and superficial tissue movement. However, in this study, one physical therapist measured the participants and intra-rater reliability was high. Finally, we did not check for shoulder or cervical spine hypomobility, which can lead to a relative hyper thoracic rotation angle. Because the thoracic spine articulates with the cervical spine and shoulders, as well as the attachment points of muscles such as the trapezius, erector spinae, serratus anterior, and other muscles.

5. Conclusion

Our results suggest that evaluations of thoracic mobility should consider the relative motions of the thoracic, lumbar, and pelvic regions, rather than the T1 angle alone. Individuals with excessive mobility in thoracic spine rotation, even those without pain, may have reduced lordotic curves in the lumbar spine or other adjacent regions. This study provides a basis for health professionals to evaluate movement dysfunctions associated with thoracic hypermobility.

Funding

This work was supported by a grant from Inje University 2023, and a grant from the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2022R1C1C2012317).

Author contributions

Concept development: SS; Design: SS; Supervision: WG; Data collection and processing: SS; Analysis and interpretation: SS; Literature search: SS; Writing: SS; Critical review: SS and WG.

Data availability

Not applicable.

Ethical approval

Ethical approval was granted by the Ethics Committee for Human Investigations at Inje University (Approval number: INJE 2020-07-016-001).

Informed consent

All participants read and signed the consent form which contained all information regarding the study. The form was approved by the University Ethics Committee for Human Investigations.

Conflict of interest

The authors declare that they have no conflict of interest.